1

Electronic Supporting Information

Helical Cobalt Borophosphates Master Durable Overall Water-Splitting

Prashanth W. Menezes,* Arindam Indra, Ivelina Zaharieva,* Carsten Walter, Stefan Loos, Stefan Hoffmann, Robert Schlögl, Holger Dau*, Matthias Driess*

a Department of Chemistry: Metalorganics and Inorganic Materials, Technische Universität Berlin, Straße des 17 Juni 135, Sekr. C2, 10623 Berlin, Germanyb Fachbereich Physik, Freie Universität Berlin, Arnimallee 14, 14195 Berlin, Germanyc Fritz Haber Institute of the Max Planck Society, Faradayweg 4–6, 14195 Berlin, Germany

Electronic Supplementary Material (ESI) for Energy & Environmental Science.This journal is © The Royal Society of Chemistry 2018

2

Fig. S1. Schematic representation of borophosphates (BPO’s, schematically by the red ball), which are intermediate compounds of systems MxOy–B2O3–P2O5–(H2O) (M = main group or transition metal).1,2 The complex anionic structures of BPO’s are comprised of interconnected trigonal–planar BO3 (ball and stick) and/or BO4 (blue tetrahedron) and PO4 (green tetrahedron) groups and their partially protonated species. The classification of BPO’s is entirely focused on the anionic partial structures, although it is clear that the cations (charge, size, and coordination behavior) have a substantial impact even on the dimensionality of the anionic structural units. In principle, the BPO’s are broadly divided into five groups as follows. 1) Tetrahedral, which comprises of anionic arrangements that contain solely BO4 and PO4 tetrahedra and no trigonal–planar borate group BO3 unit (helical BPO’s belong to this group), 2) Mixed–coordinated, that contain a BO3, and/or BO4 and PO4, 3) Metallo-, contains a 3D network structure composed of BO4, PO4, and MO4 tetrahedra (M = metal), 4) Anion–substituted, where the oxoligands of the complex anions are substituted (such as fluorine substituted), and 5) Borate–phosphates with isolated borate (BO3, BO4) and phosphate (PO4) groups.3-8

3

Fig. S2. BPO’s exclusively linked via common corners to form isolated species, oligomers, rings, chains, layers, and even frameworks. Dimensionality of complex BPO’s anionic partial structures; mixed coordinated (dark grey) and tetrahedral (light grey) BPO’s are shown in the Figure. Layers and frameworks (D = 2 – 3) are detected with compositions between B:P = 1:1 and 1:2. The oligomers, rings, and chains are known within the composition range from B:P = 6:1 to 1:4.1,2,7 The presented BPO’s falls in the category of D = 1; B:P = 1:2.

4

Fig. S3. Rietveld refinement patterns of high-resolution PXRD of the (top) LiCo(H2O)2[BP2O8]·H2O (LiCoBPO) and (bottom) NaCo(H2O)2[BP2O8]·H2O (NaCoBPO). Red dots: experimental data points; black line: calculated powder pattern; blue ticks: Bragg positions; blue line: the difference between the observed and calculated patterns.9,10

5

Fig. S4. Optical microscopic (a) and SEM images (b, c) of the single crystals of LiCoBPO. From the images, it was observed that well-defined pink crystals of hexagonal bipyramids with definite shape and facets were produced. The dimensions of the single crystals varied ranging from 0.1 mm to 0.5 mm.

Chemicals

All chemical reagents (analytical grade) were used as received without any further purification. Deionized water was used to carry out all the experiments. Commercially available cobalt(II) acetate tetrahydrate (Co(CH3COO)2·4H2O), lithium tetraborate (Li2B4O7), sodium tetraborate (Na2B4O7), phosphoric acid (H3PO4, 85%), sodium phosphate tribasic dodecahydrate (Na3PO4·12H2O), potassium hydroxide (1 M KOH, Fe < 0.05 ppm determined by ICP-AES) were obtained from Sigma Aldrich whereas ruthenium oxide (RuO2) and iridium oxide (IrO2), cobalt hydroxide (Co(OH)2), cobalt oxide (Co3O4) metallic Co was purchased from Alfa Aesar.

6

Fig. S5. Optical microscopic (a) and SEM images (b, c) of the single crystals of NaCoBPO. Although similar to LiCoBPO, well-defined pink crystals of hexagonal bipyramids with definite shape were produced, however, the roughness of the facets as well as the formation of additional facets at the edges was also found. The dimensions of the single crystals varied between 0.1 mm to 0.5 mm.

7

Fig. S6. SEM (a), and high resolution (HR) SEM images (b, c) of the single crystals of LiCoBPO. Interestingly, the hexagonal bipyramids were porous in nature and the pore sizes of each pore were approximately 1400 nm (see Fig. 1, main text)

8

Fig. S7. TEM images (a, c) and the corresponding selected area electron diffraction (SAED) (b, d) patterns of the ground crystals of LiCoBPO and the NaCoBPO. The crystals were well crystalline and also quite large in size.

9

Fig. S8. The presence of cobalt and phosphorous in LiCoBPO (top) and NaCoBPO (bottom) were determined by the EDX. The appearance of peaks for carbon and copper is due to TEM grid (carbon film on 300 mesh Cu-grid).

Table S1. Determination of chemical composition in LiCoBPO and NaCoBPO was obtained by EDX, XPS and ICP-AES analysis.

Li/Na:Co:B:P (Theo.)

Co:P (EDX)

Li/Na:Co:B:P(XPS)

Li/Na:Co:B:P(ICP-AES)

LiCo(H2O)2[BP2O8]·H2O (LiCoBPO)

1:1:1:2 ~1:2.08 1:1.04:0.94:2.1 1:1:1:2

NaCo(H2O)2[BP2O8]·H2O (NaCoBPO)

1:1:1:2 ~1:2.1 1:1.1:0.91:2.03 1:1:1:2

10

Fig. S9. IR spectra of LiCoBPO and NaCoBPO show the presence of water by characteristic absorption bands at about ~1638 cm–1 (H2O deformation) and between 3000 and 3600 cm–1 (OH stretching). In addition to this, the spectrum in each case is dominated by the strong P−O stretching modes in the region ranging from 800 to 1100 cm−1 and overlaps heavily with the strong B−O stretching vibrations in the range of 700−1200 cm−1.7,10,11

11

Fig. S10. TGA (blue solid line) and its differential (DTG, red dotted line) plot of the materials thermally treated from room temperature to 800 ˚C in a nitrogen atmosphere at the rate of 5 K/min. The dehydration of both LiCoBPO and NaCoBPO were a two-step process between 100–230 ºC and the mass loss corresponds to a distinct DTG peak at 185 and 210 ˚C. The experimental mass loss (11.04 and 9.85%) obtained in both cases were consistent with the calculated weight (11.23 and 10.65%) of two water molecules per formula unit. The second mass loss occurred at 240 to 480 ˚C with a DTG peak at 320 and 312 ˚C. The mass loss within this step was also very close to the calculated value of the release of the third water molecule evidencing the formation of the pure phase of BPOs.7

12

Fig. S11 High-resolution Co 2p XPS spectra of LiCoBPO. The Co(II) and Co(III) have almost similar 2p binding energies but can be differentiated by the Co 2p1/2−2p3/2 spin−orbit level energy spacing which is 16 eV for Co(II) and 15 eV for Co(III).12-14 The spacing of 16.2 eV and the typical satellite peaks (*) attained here is consistent with the presence of Co(II).15-18

13

Fig. S12. Typical high-resolution XPS spectra of the regions containing the (a) Li 1s (b) P 2p, (c) B 1s and (d) O1s of as-synthesized LiCoBPO. The Li 1s binding energy peaks at 55.5 eV alongside with the broad Co 3p peak at approximately 62 eV and could be directly compared to the other Li+ compounds.19,20 In the case of P 2p, the appearance of the peaks at 133.5 and 134.3 eV, corresponds to 2p3/2 and 2p1/2 that are attributed to the formation of phosphate (PO4)3- on the surface which is in good agreement with the literature reported phosphate materials.21-23 The B 1s XPS spectrum exhibited a peak at ~191.4 eV is attributed to the structure of boron linked to oxygen atoms with four coordination indicative of B3+(borate).24-26 The O 1s spectrum was deconvoluted into broad O1 and O2 peaks. The peak at ~531.4 eV (O1) is due to the large dominance of –OH species adsorbed on the surface by surface hydroxides. The O2 peak ~532.6 eV (O1) could be assigned to the chemisorbed oxygen or associated crystal water in the structure. The O1s values obtained here can be well matched with literature reported materials of phosphates and borates.21,22,26

14

Fig. S13. (a) OER CV and (b) HER LSV of LiCoBPO and NaCoBPO at a sweep rate 5 mV/s of in 1 M KOH electrolyte. The overpotential for the OER (a) was determined to be 293 mV at a current density of 10 mAcm-2 for the LiCoBPO, whereas 328 mV was obtained for the NaCoBPO. From the shape of the CVs between 0 and 0.25 V overpotential, it is also evident that a higher amount of Co3+ exists on the inner and outer surfaces of LiCoBPO compared to NaCoBPO. This Co3+ can serve as the catalytically active site for the OER for both catalysts.15-17,27 The overpotential determined for HER at -10 mAcm-2 was 245 mV for LiCoBPO and 298 mV for NaCoBPO.

15

Fig. S14. Tafel slopes derived for (a) OER and (b) HER from the polarization curves at 1 mV/s on FTO in 1 M KOH solution. A Tafel slope of 58 mVdec-1 was obtained for LiCoBPO whereas a slope of 60 mVdec-1 was recorded for NaCoBPO in OER corresponding to more favorable OER kinetics. Similarly, in HER, the Tafel slope of LiCoBPO was 98 mVdec-1, smaller than that of NaCoBPO (124 mVdec-1), indicating the more facile reaction of LiCoBPO.

16

Fig. S15. Electrochemical capacitance measurements for the estimation of the relative electrochemical active surface area (ECSA) in a non-Faradaic potential range of 0.92 V to 0.97 V vs RHE at different scan rates (10, 20, 50, 100 mV/s) for (a) LiCoBPO and (b) NaCoBPO on FTO in 1 M KOH electrolyte. (c) Estimation of double-layer capacitances (Cdl) by plotting the current density variation (Δj = (ja − jc)/2), obtained from the (a) and (b) at 0.945 V vs RHE.28,29 Nyquist plots (d) obtained from electrochemical impedance spectroscopy (EIS) for LiCoBPO and NaCoBPO. The spectra were collected with an anodic polarization potential of 1.55 V vs RHE. The curves were fitted by the inserted Randles equivalent circuit, where Rs, CPE, and Rct are the equivalent series resistance, constant phase element of the double-layer capacitance, and the charge transfer resistance, respectively.28

17

Fig. S16. The PXRD pattern of as-synthesized cobalt phosphate (see Experimental). The obtained compound could be well matched with Co3(PO4)2·8H2O that crystallizes in the monoclinic system. The cobalt phosphate was synthesized in order to have a direct comparison of the activity with the cobalt BPO catalysts.

18

Fig. S17. Comparison of (a) OER LSV and (b) HER LSV of LiCoBPO and NaCoBPO with noble commercial benchmark and as-synthesized CoPi catalysts at a sweep rate 5 mV/s in 1 M KOH electrolyte. The LiCoBPO was found to be highly active for OER resulting into lower overpotential in comparison to NaCoBPO and also surpassed the activity significantly than the RuO2, IrO2, and Co3(PO4)2·8H2O catalysts. The Pt was found to be inactive for OER within the measured range. On the other hand, Pt was extremely active for HER in comparison to LiCoBPO and NaCoBPO. The other commercial and Co3(PO4)2·8H2O catalyst was clearly less active for HER (see Table S4 for overpotentials).

19

Fig. S18. PXRD of Co3O4 (JCPDS 14-1467), Co(OH)2 (JCPDS 30-433), CoOOH (JCPDS 7-169) and the metallic Co (JCPDS 5-727) nanoparticles that were used as references for evaluating the catalytic OER and HER activities of LiCoBPO and NaCoBPO.

20

Fig. S19. Comparison of (a) OER and (b) HER polarization curves of LiCoBPO and NaCoBPO with the Co3O4, Co(OH)2, CoOOH and the metallic Co catalysts on FTO substrates at a sweep rate 5 mV/s in 1 M KOH electrolyte. For OER, the overpotential of Co3O4, Co(OH)2, CoOOH and Co catalysts at 10 mAcm-2 was 360, 382, 401 and 430 mV whereas only metallic Co could reach the current density of -10 mAcm-2 reaching an overpotential of 315 mV for HER. The attained overpotentials were significantly lower compared to the LiCoBPO catalyst (see Table S4).

21

Fig. S20. The chronoamperometric (CA) results of LiCoBPO and NaCoBPO measured under OER conditions (a) at 1.53 and 1.56 V vs RHE maintaining at 10 mAcm-2 in 1 M KOH solution. The LiCoBPO catalyst was stable for more than 24 hours and a moderate decrease in the activity for NaCoBPO was observed. The CA responses of LiCoBPO and NaCoBPO for HER (b) were measured at -0.25 and -0.3 V vs RHE (at -10 mAcm-2). Similar to the OER CA, LiCoBPO was highly stable under HER CA conditions and was better than that of NaCoBPO.

22

Fig. S21. The switchable properties during OER for LiCoBPO were measured using electrochemical conditions using 1 M KOH in a three-electrode setup. First, the pinkish FTO film (a) was subjected to OER CV (overpotential ~293 mV at 10 mAcm-2) to form a black film (b) showing immediate structural changes and subsequently, OER CA was performed for 2 h at 1.53 V vs RHE. After the OER CA, the black film was examined for HER LSV (c) forming a light brown colored film (overpotential -285 mV at -10 mAcm-2) which was then maintained for HER CA for 2 h at -0.28 V vs RHE. Remarkably, when the light brown films were further scanned for OER CV, the reversibility of the catalytic performance (d), as well as the change in color back to black, was observed. The difference in the overpotentials for the first OER CV (b) as well as the OER CV after HER (d) was observed to be negligible at 10 mAcm-2 whereas HER LSV (c) exhibited less activity than first HER LSV (see Fig. 2b).

23

Fig. S22. The switchable properties for HER for LiCoBPO were measured under electrochemical conditions using 1 M KOH in a three-electrode setup. Foremost, the pinkish FTO film (a) was subjected to HER LSV (overpotential ~-245 mV at -10 mAcm-

2) to form a light brown colored film (b) showing immediate structural changes and subsequently, HER CA was performed for 2 h at -0.25 V vs RHE. After the HER CA, the light brown film was examined for OER (c) forming a black colored film (overpotential ~300 mV at 10 mAcm-2) which was then maintained for OER CA for 2 h at 1.53 V vs RHE. Strikingly, when the black films were further investigated for HER LSV (d), the reversibility in film color was visible; however, the catalytic performance slightly poor in comparison to the first HER LSV (at 10 mAcm-2) (b). Interestingly, the overpotentials obtained for OER CV (c) could be directly compared to the first OER CV (see Fig. 2a).

24

Table S2. The comparison of OER overpotentials of LiCoBPO and NaCoBPO with other superior selected non-noble catalysts in 1M KOH.Catalyst Current density

(mAcm-2)Overpotential (mV) Mass loading

(mgcm-2) Reference

LiCoBPO/NF 10 216 ~3 This work100 324 ~3 This work1000 470 ~3 This work

NaCoBPO/NF 10 242 ~3 This work100 336 ~3 This work1000 530 ~3 This work

LiCoBPO/FTO 10 293 ~1 This workNaCoBPO/FTO 10 328 ~1 This work

CoOx@CN 10 260 ~0.42 30

Ni2P/NF 10 240 ~3 31

Ni12P5/NF 10 260 ~3 31

MoO2/NF 10 250 ~3.4 32

NiFe LDH 10 260 0.04 33

MoS2-Ni3S2 HNRs/NF 10 249 13 34

Ni3S2/NF 10 260 1.6 35

CoNi-LDH/Fe-porphyrin 10 264 0.14 36

N-G/CoO 10 340 - 37

(Ni, Co)0.85Se@NiCo-LDH 10 216 6 38

FeNi-rGO LDH 10 206 0.25 39

MoO2–CoO–C 10 270 - 40

FeOOH/Co/FeOOH 10 245 0.28 41

Co-P/Cu 10 345 1 42

CoFe-H 10 280 0.02 43

NiFe-LDH 10 240 0.19 44

Co-Fe-P 10 244 0.42 45

Co6Mo6C2/NCRGO 10 260 ~0.15 46

NixCo2x(OH)6x@Ni 10 305 4.02 47

Co5Mn-LDH/MWCNT 10 300 ~0.28 48

CoCr-LDH 10 340 0.255 49

Co- Birnessite 10 360 0.28 50

Co3ZnC 10 366 0.34 51

CoOx electrodeposited 10 380 - 52

CoSe2 10 320 0.142 53

CoMn LDH 10 324 0.22 54

Ni1-xFex NC/GC 10 330 ~2 55

Co3O4/ NiCo2O4 DSNCs 10 340 1 56

CoP/Cu 10 345 1 42

Co(OH)2 10 325 0.2 57

CoOx 10 325 2 18

Co3O4/N-rmGO 10 320 ~0.17 58

CoFeOx 10 360 - 29

NiFeOx 10 350 - 29

Co phosphide/phosphate 10 300 0.1 59

NixCo3−xO4 NWs/Ti 10 370 ~3 60

Ni-P film 10 344 1.53 61

NiCo/NS 10 334 1 62

NiCo LDH 10 367 0.23 63

25

Table S3. The comparison of HER overpotentials of LiCoBPO and NaCoBPO with other highly active selected non-noble catalysts in 1 M KOH

Catalyst Current density (mAcm-2)

Overpotential (mV) Mass loading(mgcm-2)

Reference

LiCoBPO/NF -10 121 ~3 This work-100 274 ~3 This work-1000 391 ~3 This work

NaCoBPO/NF -10 207 ~3 This work-100 307 ~3 This work-1000 506 ~3 This work

LiCoBPO/FTO -10 245 ~1 This workNaCoBPO/FTO -10 298 ~1 This work

MoS2 -10 60 - 64

MoP -10 64 3 65

Cu95Ti5 -10 60 - 66

MoC -10 77 0.76 67

NiMo -10 70 1 68

NiFe -10 94 - 69

CoMo -10 102 - 69

CoS2 -10 175 1.35 70

CoMo -10 170 454 and 579 71

CoS2 -10 145 1.7 72

CoNx -10 140 2.8 73

Mo2C -10 130 0.21 74

WS2 -10 250 0.28 75

MoB -10 225 2 76

MoC -10 130 0.21 74

Ni/Mo2C-PC -10 179 0.5 77

Cu3P/NF -10 105 1.2 78

Co9S8@NOSC-900 -10 235 5 79

CoP/CC -10 209 ~1 80

CoOX/CN -10 232 ~0.42 30

NiFe/NiFe2O4/NF -10 105 - 81

MoS2-Ni3S2 HNRs/NF -10 98 13 34

Ni-P electrodeposited -10 93 1.53 61

V/NF -10 176 0.28 82

Ni-P foam -10 135 0.5 83

NiFeP/Ni2P -10 183 - 84

NixCo3-xO4/NiCo/NiCoOx -10 155 0.7 85

NiNiP/NF -10 130 10.58 86

Ni2.5Co0.5Fe/NF -10 150 0.25 87

Ni2P/GC -20 250 0.38 88

Ni5P4 film -10 180 - 89

NiCo2S4 -20 194 4 90

CoNiP/NF -10 155 1 91

Ni3FeN-NPs -10 158 0.2 92

Zn0.76Co0.24S/CoS on Ti -10 200 1 93

Ni3S2@Ni -10 195 - 94

NiS/NiF -10 150 43 95

Co9S8−NixSy/NF -10 163 7 96

Ni2.3%CoS2/CC -10 150 ~1 97

26

Table S4. The comparison of OER and HER overpotentials of BPO’s with other benchmark catalysts tested using our three-electrode set-up on FTO and NF in 1 M KOH.

Catalyst Current density (mAcm-2)

OER overpotential (mV)

Current density (mAcm-2)

HER overpotential (mV)

On FTO (loading ~1 mgcm-2)LiCoBPO 10 293 -10 245 NaCoBPO 10 328 -10 298 Co3(PO4)2·8H2O 10 430 -10 358 IrO2 10 400 -10 430 RuO2 10 354 -10 -Pt wire 10 - -10 42 FTO 10 - -10 -

On NF (loading ~3 mgcm-2)LiCoBPO 10 216 -10 121

100 324 -100 274NaCoBPO 10 242 -10 207

100 336 -100 307Pt 10 - -10 42

100 - -100 130IrO2 10 310 -10 220

100 430 -100 335RuO2 10 292 -10 229

100 420 -100 370NF 10 490 -10 260

100 - -100 470

27

Fig. S23. Grazing Incidence X-ray diffraction patterns (GIXRD) of LiCoBPO, post-OER CV and OER CA films along with bare FTO as a reference. Although no major structural changes in the diffraction pattern were observed after OER, the intensity of the reflection at ~11º was decreased in the post OER samples with an unassignable peak at ~28º. The GIXRD result shows that the core of the LiCoBPO remained very crystalline under OER conditions whereas the surface of the particles transformed into the amorphous structure (which cannot be identified by PXRD and thus, evidenced by HR-TEM analysis).

28

Fig. 24. GIXRD of LiCoBPO, post HER CV and HER CA films along with bare FTO as a reference. Similar to OER CA, no major structural changes in the diffraction pattern was observed after OER, apart from the reflection at ~11º which was decreased in the post HER samples with an unassignable peak at ~28º. The GIXRD result shows that the core of the LiCoBPO remained very crystalline under HER conditions apart from the surface passivation (as concluded from the HR-TEM analysis).

29

Fig. S25. FT-IR transmission spectrum of as-prepared LiCoBPO and the LiCoBPO after OER CA experiments. The bands at 3320 cm-1 showed that the LiCoBPO catalyst after OER is largely hydroxylated.17

30

Fig. S26. The presence of cobalt and phosphorous in LiCoBPO after OER CA measurements were determined by the EDX analysis. The appearance of peaks for copper is due to the TEM grid (carbon film on 300 mesh Cu-grid) and the potassium is from the used electrolyte under electrochemical conditions.

31

Fig. S27. The presence of cobalt and phosphorous in LiCoBPO after HER CA measurements were determined by the EDX analysis. The appearance of peaks for copper is due to the TEM grid (carbon film on 300 mesh Cu-grid) and the potassium is from the used electrolyte under electrochemical conditions.

32

Fig. S28. TEM (20 nm) and HRTEM (5 nm) images of the surface of the thin films of LiCoBPO after OER CV (a, b) and after OER CA for 24 h (c, d) measurements in 1 M KOH solution. In the case of OER CV, the modification is very limited that means a very thin amorphous shell (~2 nm) started growing on the surface of the LiCoBPO particles indicating the possible formation of hydroxylated phase. After OER CA, a very thick amorphous shell (> 20 nm) was observed indicating a major surface structural change during OER. This type of amorphous shell on the surface of the particles indicating the formation of hydroxylated or oxy-hydroxylated phases has been already discussed in the literature reported transition metal-based catalysts.17,18,31,98 The corresponding SAED pattern are represented in the inset.

33

Fig. S29. TEM (50 nm) and HRTEM (5 nm) images of the surface of the thin films of LiCoBPO after HER LSV (a, b) and after HER CA for 24 h (c, d) measurements in 1 M KOH solution. In the case of both HER LSV and HER CA, no indication of the surface shell was observed. The particles were well crystalline (see b inset) but the surface was also passivated with CoOx as shown in (d).31 The SAED’s are shown in the inset.

34

Fig. S30. The difference in the Co 2p1/2−2p3/2 spin−orbit level energy spacing in the XPS spectra of as-prepared LiCoBPO, OER CV, OER CA, HER LSV, and HER CA, respectively. For more details, refer to Fig. S31 and S32.

35

Fig. S31. The Co 2p XPS spectra of LiCoBPO after (a) OER CV and (b) OER CA. The Co(II) and Co(III) sites have almost similar 2p binding energies but ultimately, they can be differentiated by the Co 2p1/2−2p3/2 spin−orbit level energy spacing.12-14 This difference is 16.0 eV for high-spin Co(II) and 15.0 eV for low-spin Co(III). In the case of as-synthesized LiCoBPO, the spin−orbit level energy spacing was 16.2 eV indicating that the oxidation state of Co is ~+2. However, after the OER CV, this difference was reduced to 15.6 eV indicating some of the surface oxidation of Co(II) to Co(III). However, after OER CA, the spin−orbit level energy spacing was found to be 15.1 eV that shows almost complete surface oxidation to Co(III). The results obtained here are consistent with the literature reported materials with Co(II) and Co(III).12,16-18 Interestingly, the surface oxidation of cobalt could also be seen from the HR-TEM results where a thick amorphous shell formation was observed. In both (a) OER CV and (b) OER CA, the 2p3/2 deconvoluted peaks at binding energy of ~780 and ~781.8 eV could directly be attributed to the Co(III) and Co(II) of the literature reported materials.12 Similarly, the 2p1/2 deconvoluted peaks at binding energy of ~795 and ~797 eV are also consistent with the presence of Co(III) and Co(II). In addition, for OER CV, four satellite peaks (represented by *) were also observed at ~785 and ~ 790.5 eV for 2p3/2 and at 804 eV for 2p1/2.

99 Furthermore, in the case of OER CA, only a single satellite peak ~782.5 eV was observed which also signifies the surface oxidation of cobalt.16,17 From detailed analysis and comparing the deconvoluted areas of Co(III) and Co(II) between OER CV and OER CA, it was evident that Co(III) peaks of both 2p3/2 (red curve) and 2p1/2 (green curve) were increased after CA while the Co(II) curves (blue and orange curves) were decreased. The resulting percentage of the area of Co 2p region is shown in Table S5.

36

Table S5. The distribution of the area in the region of Co 2p3/2 and Co 2p1/2 XPS with respect to Co(II) and Co(III) oxidation states and satellites for LiCoBPO after OER CV and OER-CA experiments.

Binding Energy (eV) OER-CV area in % OER-CA area in %Co 2P3/2 Co(III) ~780.5 5.02 15.72Co 2P1/2 Co(III) ~795.0 2.48 7.81Co 2P3/2 Co(II) ~781.8 37.97 40.34Co 2P1/2 Co(II) ~797.5 18.79 20.04Co 2P Satellite 1 ~784.3 21.17 16.09Co 2P Satellite 2 ~789.5 4.93 -Co 2P Satellite 3 ~804.5 9.64 -

Fig. S32. The Co 2p XPS spectra of LiCoBPO after (a) HER LSV and (b) HER CA. As described for the OER, the Co 2p1/2−2p3/2 spin−orbit level energy spacing obtained for HER LSV and HER CA was 16.0 and 15.8 eV, respectively, indicating that there was not much of a difference in the +2 oxidation state of cobalt from the as-synthesized LiCoBPO.12-14 Both in the case of HER LSV and HER CA, the 2p3/2 deconvoluted peaks at a binding energy of ~781.8 eV suggests that the highest contribution on the surface is from the Co(II) 99. Similarly, the smallest peaks at ~780 correspond to Co(III) that may have formed from the slight surface passivation. Furthermore, the 2p1/2 deconvoluted peaks at a binding energy of ~797 eV are also consistent with the presence of Co(II) as major phase and the small peak at ~795 eV corresponds to the surface passivation forming possibly Co(III) species. The results obtained here are consistent with the literature reported examples and can be well corroborated with the presented Table S6.15,16,18 The values within the Table show that there were no drastic changes in the oxidation state of Co before and after electrochemical measurements. Although, a large amount of metallic Co contribution could be observed in in-situ XAS measurement (see later on), however, it was not possible to determine Co(0) in XPS as it is quickly re-oxidizes to Co(II) on the surface of the crystals when exposed to ambient conditions.

37

Table S6. The distribution of the area in the region of Co 2p3/2 and Co 2p1/2 XPS with respect to Co(II) and Co(III) oxidation states and satellites for LiCoBPO after HER LSV and HER CA experiments.

Binding Energy (eV) HER-LSV area in % HER-CA area in %Co 2P3/2 Co(III) ~ 779.7 9.0 11.89Co 2P1/2 Co(III) ~ 794.8 4.77 5.91Co 2P3/2 Co(II) ~ 780.5 46.26 46.58Co 2P1/2 Co(II) ~ 796.5 22.98 23.14Co 2P Satellite 1 ~ 782.3 16.40 12.50

Fig. S33. Typical high-resolution Li 1s XPS spectra of LiCoBPO (green), LiCoBPO after OER CV (red) and LiCoBPO after HER LSV (violet). The Li 1s binding energy peaks at 55.5 eV alongside with the broad Co 3p peak at approximately 62 eV confirms that Li is still present in the surface of the material and values here could be directly compared to other Li+ compounds.19,20 Interestingly, after the long-term stability measurement, no lithium was detected on the surface implying a loss of lithium and a complete change of near-surface structure in both HER and OER conditions under applied potentials.

38

Fig. S34. Typical high-resolution P 2p XPS spectra of LiCoBPO after electrochemical OER CV, OER CA, HER LSV, and HER CA. For comparison, the as-prepared LiCoBPO is also shown. The peaks at binding energy ~133.8 eV clearly indicated the presence of the only phosphate (PO4

3-) on the surface in all investigated materials of HER and OER, which is in good agreement with the literature reported phosphate materials.21-

23

39

Fig. S35. High-resolution B 1s XPS spectra of LiCoBPO after electrochemical OER CV, OER CA, HER LSV, and HER CA. For comparison, the as-prepared LiCoBPO is also shown. The B 1s XPS spectra exhibited the main peak at ~191.4 eV in both OER and HER electrochemical conditions corresponding to the structure of boron linked to oxygen atoms with four coordination indicative of B3+ (borate) evidencing boron did not change its coordination.24-26 In CA treated samples, an additional peak starts to appear at 191.4 eV, which could be attributed to the formation of B2O3 on the surface.100,101

40

Fig. S36. O 1s XPS spectra of LiCoBPO after electrochemical (a) OER CV, (b) OER CA, (c) HER LSV and (d) HER CA. In the case of (a) and (c), the O1s spectrum can be deconvoluted into broad O1, O2, and O3 peaks. The peak at ~531.4 eV (O1) is due to the large dominance of –OH species adsorbed on the surface by surface hydroxides especially after the electrochemical OER treatment. The O2 peak ~532.6 eV (O1) could be assigned to the chemisorbed oxygen or the crystal water associated with the structure. Interestingly, O3 peak between ~529 and ~530 eV indicates the formation of oxides on the surface. However, for (b), O3 peak was shifted towards higher binding energies suggesting the formed metal oxides further hydroxylates under CA conditions in strongly alkaline media. The values obtained here can be easily matched with similar literature reported materials such as phosphates, borates and oxides.16,18,21,26,99,102

41

Table S7. Averaged results of the four-point probe resistivity measurement of the as-synthesized LiCoBPO and NaCoBPO as well as the after catalytic experiments. Although the as-prepared materials were conductive, the resistivity of the LiCoBPO films decreased significantly post electrocatalytic OER and HER measurements improved charge transfer through the materials.

Catalyst Resistivity (Ω/sq)FTO 7.14

LiCoBPO/FTO 2.53×108

After OER CA 3.78×106

After HER CA 2.33×106

NaCoBPO/FTO 4.31×108

42

Table S8. Averaged Co oxidation state determined from the bond-valence-sum of LiCoBPO.

Compound Co-oxidation statePowder 2.04as deposited 1.83HER LSV 1.86HER CA 1.18OER CV 2.42OER CA 2.43

Table S9. Simulation of the material after exposure to oxidizing potentials as a linear combination of the as-deposited LiCoBPO and CoCat. For simulation were used k3-weighted experimental spectra in the range 3 to 14 Å-1. In the simulation, the simulated spectrum of the as-deposited LiCoBPO was used (meaning that not the experimental, but calculated spectrum was used, taking both distances and coordination numbers as determined for the as-deposited LiCoBPO; parameters given in Table S11). For the CoCat spectrum, the simulation parameters from the CoCat spectrum were used (Table S12). In addition, a multiple scattering shell for the collinearly aligned three Co atoms connected by di--oxo bridge was used. The distance for this shell was fixed to the doubled short Co-Co distance in the CoCat (2*2.81 Å). The coordination number (N) gives information about the degree of long-range order in the oxide (for an ideal infinite oxide layer it should be equal to 6) and was determined by a simulation. It is shown in the third column of the table. All Debye-Waller factors were fixed to 0.063 Å.

OER LiCoBPO, % CoCat, % Co-CoMS (N)after CVs 63.0 ± 4.4 37.0 ± 4.4 1.2 ± 0.8after CA 62.5 ± 4.4 37.5 ± 4.4 1.1 ± 0.7

Table S10. Simulation of the material after exposure to oxidizing potentials as a linear combination of the as-deposited LiCoBPO, metallic Co and CoO. For simulation k3-weighted experimental spectra were used in the range 3 to 14 Å-1. In the simulation, the simulated spectrum of the as-deposited LiCoBPO was used (parameters in Table S11). For the metallic spectrum, the simulation parameters from the simulation of the experimental metallic Co were used (Table S14). For the CoO, we did not have an experimental spectrum and we used three shells (one Co and two O) with coordination numbers predicted by the crystal structure of CoO and distances varied during the simulations. These parameters are shown in Table S14. All Debye-Waller factors were fixed to 0.063 Å.

HER LiCoBPO, % metallic Co, % CoO, %after LSVs 83.4 ± 4.5 0.4 ± 2.1 16.2 ± 6.6after CA 49.7 ± 3.9 38.3 ± 2.0 12.0 ± 5.9

43

Table S11. Simulation parameters for the LiCoBPO powder and as-deposited material. For simulation were used k3-weighted experimental spectra in the range 3 to 14 Å-1. The errors (grey color) represent the 68% confidence interval of the respective fit parameter (R, absorber-backscatter distance; σ, Debye-Waller parameter), which does not cover systematic errors due to imperfect EXAFS phase functions. The coordination numbers (N, XRD) and distances (R, XRD) calculated according to the crystal structure are shown in blue for comparison and not used in the simulation. The XAS data was collected at 20 K whereas the XRD data was collected at room temperature; this temperature distance is predicted to result in slightly shorter distances in the low-temperature EXAFS data. The Fourier-filtered error factor was calculated using a Fourier filter between 1 and 5 Å resulting in a value of 28.3% for the powder and 21.8% for the as-deposited material. The Debye-Waller factors for all shells were fixed to 0.063 Å.

Co-O Co-P Co-O Co-O Co-ON, XRD 6 4 6 4 11N, powder 5.7 2.3 5.0 4.7 11.6error 0.6 1.0 2.8 3.9 5.3N, as deposited 5.2 2.8 3.7 5.7 9.1error 0.4 0.7 2.0 2.3 5.4R, powder [Å] 2.08 3.20 3.67 4.10 4.56error 0.01 0.02 0.04 0.05 0.03R, as deposited [Å] 2.08 3.27 3.67 4.06 4.62error 0.01 0.02 0.03 0.03 0.04R, XRD [Å] 2.10 3.30 3.71 4.11 4.47

Table S12. Simulation parameters for the CoCat spectrum taken from Ref. 9 (Risch et al.). For simulation were used k3-weighted experimental spectra in the range 3 to 14 Å-1. The errors (grey color) represent the 68% confidence interval of the respective fit parameter (R, absorber-backscatter distance; σ, Debye-Waller parameter). The Fourier-filtered error factor was calculated using a Fourier filter between 1 and 5 Å resulting in a value of 17.5%. The asterisk (*) indicate the parameters fixed during the simulation.

Co-O Co-CoN 6* 3.5error - 0.5R [Å] 1.89 2.81*error 0.01 -s [Å] 0.060 0.0632*error 0.007 -

44

Table S13. Simulation parameters for the spectrum of the metallic Co. For simulation were used k3-weighted experimental spectra in the range 3 to 14 Å-1. The errors (grey color) represent the 68% confidence interval of the respective fit parameter (R, absorber-backscatter distance; σ, Debye-Waller parameter). The Fourier-filtered error factor was calculated using a Fourier filter between 1 and 5 Å resulting in a value of 18.7%. The asterisk (*) indicate the parameters fixed during the simulation (all coordination numbers were fixed to their values predicted by the structure of the metallic Co). The distances (R, XRD) calculated according to the crystal are shown in blue for comparison.

Co-Co Co-Co Co-Co Co-Co Co-CoN 12* 6* 2* 18* 18*R [Å] 2.50 3.54 4.01 4.36 4.86error 0.00 0.00 0.02 0.00 0.00R, XRD [Å] 2.50 3.54 4.07 4.34 4.78s [Å] 0.046 0.036 0.055* 0.058 0.033error 0.001 0.004 <0.001 0.003 0.003

Table S14. Simulation parameter obtained by the simulation of the LiCoBPO exposed to reducing potentials (Table S10) which were assigned to a CoO phase. In the simulation, the coordination numbers (N) of the shells added to represent the CoO phase were fixed to the values corresponding to the crystal structure of CoO (indicated in the table with asterisks, *). The interatomic distances (R) were varied during the simulation. The errors (grey color) represent the 68% confidence interval. The predicted from the CoO structure distances (R, XRD) are shown for a comparison in blue color. The Debye-Waller parameters were all fixed to 0.063 Å.

Co-O Co-Co Co-ON 4* 12* 12*R, XRD [Å] 1.97 3.21 3.78R [Å] 1.93 3.18 3.75error 0.05 0.01 0.10

45

Fig. S37. (a) OER LSV and (b) HER LSV of LiCoBPO and NaCoBPO deposited on NF at a sweep rate of 5 mV/s of in 1 M KOH electrolyte. The overpotential for the OER (a) was determined to be 216 mV at a current density of 10 mAcm-2 for the LiCoBPO whereas a slightly higher overpotential of 242 mV was obtained for the NaCoBPO. At a current density of 100 mAcm-2, the overpotentials were quite close with ~324 and ~336 mV for both LiCoBPO and NaCoBPO, respectively. The overpotential acquired for HER at -10 mAcm-2 was 121 mV for LiCoBPO and 207 mV for NaCoBPO. Similarly, at -100 mAcm-2, the attained overpotential was 274 and 307 mV for LiCoBPO and NaCoBPO, respectively. The overpotential of LiCoBPO in both cases was extremely low in comparison to the transitional metal-based catalyst systems in alkaline media (see Table S3 and S4).

46

Fig. S38. Tafel slopes derived for (a) OER and (b) HER from the polarization curves at 1 mV/s on NF in 1 M KOH solution. A Tafel slope of 62 mVdec-1 was attained for LiCoBPO while a higher Tafel slope of 99 mVdec-1 was obtained for NaCoBPO showing better OER kinetics of LiCoBPO. In HER, the Tafel slope of LiCoBPO was 121 mVdec-1 which was smaller than NaCoBPO (128 mVdec-1).

47

Fig. S39. Electrochemical capacitance measurements for the estimation of the relative ECSA in a non-Faradaic potential range of 0.92 V to 1.02 V vs RHE at different scan rates (10, 20, 50, 100 mV/s) for (a) LiCoBPO and (b) NaCoBPO on NF in 1 M KOH electrolyte. (c) Estimation of double-layer capacitances (Cdl) by plotting the current density variation (Δj = (ja − jc)/2), obtained from the (a) and (b) at 0.97 V vs RHE.34,103 Nyquist plots (d) obtained from electrochemical impedance spectroscopy (EIS) for LiCoBPO and NaCoBPO. The spectra were collected with an anodic polarization potential of 1.5 V vs RHE. The inset in (d) is the enlarged EIS curves.104

48

Fig. S40. Comparison of (a) OER LSV and (b) HER LSV of LiCoBPO and NaCoBPO with noble commercial benchmark on NF at a sweep rate of 5 mV/s of under 1 M KOH electrolyte. In OER, both LiCoBPO and NaCoBPO surpassed the activity substantially to IrO2 and the Pt was found to be inactive for OER within the measured range. On the other hand, Pt was extremely active for HER in comparison to LiCoBPO and NaCoBPO and IrO2 was clearly less active (see Table S4 for overpotentials).

49

Fig. S41. Comparison of (a, b) OER and (b) HER polarization curves of LiCoBPO and NaCoBPO with the Co3O4, Co(OH)2, CoOOH and the metallic Co catalysts on NF. substrates at a sweep rate 5 mV/s in 1 M KOH electrolyte. On FTO, the overpotential of Co3O4, Co(OH)2, CoOOH and Co catalysts at 10 mAcm-2 was 360, 380, 400 and 430 mV whereas, for HER, at the same current density the overpotential achieved was 217, 234, 222 and 210 mV, respectively. The attained overpotentials were significantly lower compared to the LiCoBPO catalyst (see Table S4).

50

Fig. S42. The chronoamperometric (CA) results of LiCoBPO and NaCoBPO on NF measured in (a) OER conditions at 1.45 and 1.48 V vs RHE maintaining at 10 mAcm-2 in 1 M KOH solution. The LiCoBPO catalyst was stable for more than 40 hours but a slight decrease in the NaCoBPO was seen. The CA responses of LiCoBPO and NaCoBPO in (b) HER conditions measured at -0.13 and -0.21 V (at -10 mAcm-2) showed activation over time.

51

Fig. S43. The SEM images of the as-prepared LiCoBPO on NF (a, b) displaying closely packed crystallites at the electrode substrate (c, d).

52

Fig. S44. The SEM images of the LiCoBPO on NF after OER CA (a, b). The images showed mostly the surface modified particles (c, d) confirming dramatic changes during electrochemical experiments and are responsible for enhancing the catalytic activity.

53

Fig. S45. Overall water-splitting with LiCoBPO║LiCoBPO on nickel foam in 1 M KOH solution with a two-electrode set-up at a potential of 1.53 V (at a current of 10 mAcm-2) was applied. Vigorous bubble and gas evolution was observed.

54

Fig. S46. Comparison of the polarization curves of bare NF∥NF, LiCoBPO/NF∥LiCoBPO/NF and NaCoBPO/NF∥NaCoBPO/NF with the precious metal-based Pt (-)∥IrO2/NF(+) and Pt (-)∥RuO2/NF(+) for overall water-splitting in 1 M KOH at a scan rate of 5 mV/s.

55

Fig. S47. The determination of Fe impurities from ICP-AES analysis of bare NF, as-prepared LiCoBPO/NF, anode and cathode used for overall water-splitting (73 d). The results showed that the amount Fe present in the NF was almost similar to that the other samples that rule out the substantial contribution of Fe from the 1 M KOH electrolyte.

Table S15. ICP-AES analysis of the anode and the cathode after 73 days of electrolysis to determine the compositional changes of LiCoBPO. The significant loss of B and P along with the Li shows that major structural modification (CoOOH) takes place at the surface. This can also be correlated to the gradual increase in current density in the CA experiment.

Theoretical Ratio (Co : B : P : Li)

Ratio by ICP(Co : B : P : Li)

LiCoBPO 1 : 1 : 2 : 1 1.0 : 1.02 : 2.0 : 0.98LiCoBPO/Anode - 1.0 : 0.55 : 0.85 : 0.06LiCoBPO/Cathode - 1.0 : 0.80 : 0.20 : 0.06

56

Fig. S48. To rule out the possible influence or contribution of NF in alkaline overall water-splitting, two-electrode set up was fabricated with (a) bare NF∥NF at a potential of 1.53 V (cell potential of LiCoBPO at 10 mAcm-2) as well as at 1.84 V (cell potential of NF at 10 mAcm-2) and the change in the current was monitored. A continuous decrease in the current density with respect to time was attained in both cases revealing the negligible contribution of NF in overall catalysis in 1 M KOH solution. In addition to this, the ICP-AES of the electrolyte was also measured that showed less than 0.001% of Ni leaching. This further eliminates the incorporation of nickel into the structure as well as the possible formation of Ni(OH)2.

57

ESI Note 1

Based on the microscopic and spectroscopic results, the increase in current density over time (up to 40 days) during the reaction of overall water-splitting of LiCoBPO on NF can be attributed to:

1. Easy separation of low coordinated Li ions from the crystal structure of LiCoBPO under strongly alkaline electrochemical conditions forming a vacanted, defected and disordered structure to accelerate catalysis both at cathode and anode.

2. Continuous surface transformation of LiCoBPO formation of the Co-rich amorphous shell with growing layer thickness at the anode.

3. A slower detachment of B and P from the structure (see Table S15) to form more vacancies at both at anode and cathode and this accommodates further structural transformations to expose more active sites to facilitate overall water-splitting.

After the 40 days of catalysis, no further increase in current was observed. This means that the LiCoBPO allows maximum transformation and reorganization of the structure at the surface of anode and cathode to form highly active Co-rich phase. In addition, the inner core of the catalyst still contains LiCoBPO (see Table S15) that signifies the crucial roles of B and P in reorganizing the structure, thereby improving overall water-splitting efficiency.

58

Calculation of Faradaic efficiency

The Faradaic efficiency (FE) of LiCoBPO in 1M KOH towards oxygen and hydrogen evolution reaction was measured in a two-electrode configuration where nickel foam loaded with LiCoBPO were used as both cathode and anode in a closed electrochemical cell. The electrolyte and cell were first degassed with Argon for 30 min under stirring. Afterward, the constant current density of 10 mAcm-2 was applied for a certain period of time. At the end of electrolysis, the gaseous samples were drawn from the headspace by a gas-tight syringe and analyzed by a GC calibrated for H2, and O2. Each injection was repeated at least three times and the average value is presented.

The Faradaic efficiency (FE) is calculated based on:

VH2, VO2 is the evolved volume of hydrogen and oxygen, F is the Faraday constant (96485.33289 C/mol), Vm is the molar volume of the gas, j is the current density (10 mAcm-2) and t is the time period of electrolysis.

Table S16: Calculation of Faradaic efficiencyj /mAcm-2 t/

secVH2/ mL

VO2/ mL

VH2:VO2 FE (H2,%)

FE (O2,%)

LiCoBPO 10 900 1.04 0.51 2:1 99 97

After 73 d 10 900 1.02 0.48 2:1 98 94

59

Fig. S49. The electrolysis was performed in a modified two-electrode LiCoBPO/NF║LiCoBPO/NF configuration in 1 M KOH at a constant current density of 10 mAcm-2 was carried out in an inverted graduated electrochemical cell to allow collection of H2 and O2 separately at atmospheric pressure as shown (left). The initial level of the electrolyte was noted and then the valves were closed (top right). During electrolysis, as a result of evolution and collection of H2 and O2 at the upper part of the cell, the level of electrolyte goes down and the change in volume over time is recorded (Figure right). The ratio of volumes of H2 and O2 remained almost ~2:1 over the period of electrolysis. The evolved gases H2 and O2 were also identified by a gas-chromatograph.

60

Fig. S50. Graph displaying the volume change as a result of H2 and O2 evolution of the experiment carried out in Fig. S49. The ratios of H2 and O2 were obtained from the modified two electrodes LiCoBPO║LiCoBPO on NF as both cathode and anode in 1 M KOH solution at a current of 10 mA for 1 h. The obtained ratios of H2 was approximately twice larger than the O2 demonstrating the effective selectivity and reactivity of the catalysts.

61

Fig. S51. (a) polarization curve of LiCoBPO/FTO∥LiCoBPO/FTO fabricated via a two-electrode setup in 1 M KOH at a scan rate of 5 mV/s displaying a cell potential of 1.94 at a current density of 10 mAcm-2 and (b) CA curves at a potential of 1 95 V over 10 days. Similar to LiCoBPO/NF substrate, the CA of LiCoBPO/FTO substrate which also showed continuous growth in current density over time confirming the fast and easy separation of Li and transformation of the material into a Co-rich surface (see Note 1 for more details) as well as excluding the effect of the electrode substrate.

62

References:

1. R. Kniep, H. Engelhardt and C. Hauf, Chem. Mater. , 1998, 10, 2930-2934.2. B. Ewald, Y. X. Huang and R. Kniep, Z. Anorg. Allg. Chem., 2007, 633, 1517-1540.3. B. Ewald, Y. Prots, P. Menezes, S. Natarajan, H. Zhang and R. Kniep, Inorg. Chem.,

2005, 44, 6431-6438.4. P. W. Menezes, S. Hoffmann, Y. Prots and R. Kniep, Z. Kristallogr.-NCS, 2009, 224,

1-2.5. P. W. Menezes, S. Hoffmann, Y. Prots and R. Kniep, Z. Kristallogr.-NCS, 2006, 221,

251-252.6. P. W. Menezes, S. Hoffmann, Y. Prots and R. Kniep, Z. Kristallogr.-NCS, 2006, 221,

253-254.7. P. W. Menezes, Technical University of Dresden, 2009.8. P. W. Menezes, S. Hoffmann, Y. Prots, W. Schnelle and R. Kniep, Inorg. Chim. Acta

2010, 363, 4299-4306.9. P. W. Menezes, S. Hoffmann, Y. Prots and R. Kniep, Z. Kristallogr.-NCS, 2007, 222,

1-2.10. P. W. Menezes, S. Hoffmann, Y. Prots and R. Kniep, Z. Kristallogr.-NCS, 2008, 223,

333-334.11. R. Kniep, H. G. Will, I. Boy and C. Rohr, Angew. Chem. Int. Ed., 1997, 36, 1013-

1014.12. M. Oku and K. Hirokawa, J. Electron Spectros. Relat. Phenom., 1976, 8, 475-481.13. D. C. Frost, C. A. McDowell and I. S. Woolsey, Mol. Phys., 1974, 27, 1473-1489.14. C. A. Kent, J. J. Concepcion, C. J. Dares, D. A. Torelli, A. J. Rieth, A. S. Miller, P.

G. Hoertz and T. J. Meyer, J. Am. Chem. Soc., 2013, 135, 8432-8435.15. P. W. Menezes, A. Indra, N. R. Sahraie, A. Bergmann, P. Strasser and M. Driess,

ChemSusChem, 2015, 8, 164-171.16. P. W. Menezes, A. Indra, D. Gonzalez-Flores, N. R. Sahraie, I. Zaharieva, M.

Schwarze, P. Strasser, H. Dau and M. Driess, ACS Catal., 2015, 5, 2017-2027.17. P. W. Menezes, A. Indra, A. Bergmann, P. Chernev, C. Walter, H. Dau, P. Strasser

and M. Driess, J. Mater. Chem. A, 2016, 4, 10014-10022.18. A. Indra, P. W. Menezes, C. Das, C. Gobel, M. Tallarida, D. Schmeisser and M.

Driess, J. Mater. Chem. A, 2017, 5, 5171-5177.19. S. Verdier, L. El Ouatani, R. Dedryvere, F. Bonhomme, P. Biensan and D. Gonbeau,

J. Electrochem. Soc., 2007, 154, A1088-A1099.20. A. W. Moses, H. G. G. Flores, J. G. Kim and M. A. Langell, Appl Surf. Sci., 2007,

253, 4782-4791.21. M. Pramanik, C. L. Li, M. Imura, V. Malgras, Y. M. Kang and Y. Yamauchi, Small,

2016, 12, 1709-1715.22. F. H. Saadi, A. I. Carim, E. Verlage, J. C. Hemminger, N. S. Lewis and M. P.

Soriaga, J. Phys. Chem. C, 2014, 118, 29294-29300.23. Z. G. Zhao, J. Zhang, Y. Y. Yuan, H. Lv, Y. Y. Tian, D. Wu and Q. W. Li, Sci. Rep.,

2013, 3.24. S. K. Choi, W. Choi and H. Park, Phys. Chem. Chem. Phys., 2013, 15, 6499-6507.25. Y. J. Wang and M. Trenary, Chem. Mater. , 1993, 5, 199-205.

63

26. X. Y. Sun, Y. X. Ding, B. S. Zhang, R. Huang and D. S. Su, Chem. Commun., 2015, 51, 9145-9148.

27. P. W. Menezes, A. Indra, O. Levy, K. Kailasam, V. Gutkin, J. Pfrommer and M. Driess, Chem. Commun., 2015, 51, 5005-5008.

28. Q. Kang, L. Vernisse, R. C. Remsing, A. C. Thenuwara, S. L. Shumlas, I. G. McKendry, M. L. Klein, E. Borguet, M. J. Zdilla and D. R. Strongin, J. Am. Chem. Soc., 2017, 139, 1863-1870.

29. C. C. L. McCrory, S. H. Jung, J. C. Peters and T. F. Jaramillo, J. Am. Chem. Soc., 2013, 135, 16977-16987.

30. H. Y. Jin, J. Wang, D. F. Su, Z. Z. Wei, Z. F. Pang and Y. Wang, J. Am. Chem. Soc., 2015, 137, 2688-2694.

31. P. W. Menezes, A. Indra, C. Das, C. Walter, C. Gobel, V. Gutkin, D. Schmeisser and M. Driess, ACS Catal. , 2017, 7, 103-109.

32. Y. S. Jin, H. T. Wang, J. J. Li, X. Yue, Y. J. Han, P. K. Shen and Y. Cui, Adv. Mater., 2016, 28, 3785-3790.

33. B. M. Hunter, J. D. Blakemore, M. Deimund, H. B. Gray, J. R. Winkler and A. M. Muller, J. Am. Chem. Soc., 2014, 136, 13118-13121.

34. Y. Q. Yang, K. Zhang, H. L. Ling, X. Li, H. C. Chan, L. C. Yang and Q. S. Gao, ACS Catal. , 2017, 7, 2357-2366.

35. L. L. Feng, G. T. Yu, Y. Y. Wu, G. D. Li, H. Li, Y. H. Sun, T. Asefa, W. Chen and X. X. Zou, J. Am. Chem. Soc., 2015, 137, 14023-14026.

36. C. Zhang, J. W. Zhao, L. Zhou, Z. H. Li, M. F. Shao and M. Wei, J. Mater. Chem. A, 2016, 4, 11516-11523.

37. S. Mao, Z. H. Wen, T. Z. Huang, Y. Hou and J. H. Chen, Energy Environ. Sci., 2014, 7, 609-616.

38. C. Xia, Q. Jiang, C. Zhao, M. N. Hedhili and H. N. Alshareef, Adv. Mater., 2016, 28, 77-+.

39. X. Long, J. K. Li, S. Xiao, K. Y. Yan, Z. L. Wang, H. N. Chen and S. H. Yang, Angew. Chem. Int. Ed., 2014, 53, 7584-7588.

40. B. B. Li, Q. Liang, X. J. Yang, Z. D. Cui, S. Z. Qiao, S. L. Zhu, Z. Y. Li and K. Yin, Nanoscale, 2015, 7, 16704-16714.

41. J. X. Feng, H. Xu, Y. T. Dong, S. H. Ye, Y. X. Tong and G. R. Li, Angew. Chem. Int. Ed., 2016, 55, 3694-3698.

42. N. Jiang, B. You, M. L. Sheng and Y. J. Sun, Angew. Chem. Int. Ed., 2015, 54, 6251-6254.

43. W. Liu, H. Liu, L. N. Dang, H. X. Zhang, X. L. Wu, B. Yang, Z. J. Li, X. W. Zhang, L. C. Lei and S. Jin, Adv. Func. Mater., 2017, 27.

44. K. Zhang, W. H. Wang, L. Kuai and B. Y. Geng, Electrochimi. Acta, 2017, 225, 303-309.

45. T. Zhang, J. Du, P. X. Xi and C. L. Xu, ACS Appl. Mater. Interfaces, 2017, 9, 362-370.

46. Y. J. Tang, C. H. Liu, W. Huang, X. L. Wang, L. Z. Dong, S. L. Li and Y. Q. Lan, ACS Appl. Mater. Interfaces, 2017, 9, 16978-16986.

47. Z. Q. Liu, G. F. Chen, P. L. Zhou, N. Li and Y. Z. Su, J. Power Sources, 2016, 317, 1-9.

64

48. G. Jia, Y. F. Hu, Q. F. Qian, Y. F. Yao, S. Y. Zhang, Z. S. Li and Z. G. Zou, ACS Appl. Mater. Interfaces, 2016, 8, 14527-14534.

49. C. L. Dong, X. T. Yuan, X. Wang, X. Y. Liu, W. J. Dong, R. Q. Wang, Y. H. Duan and F. Q. Huang, J. Mater. Chem. A, 2016, 4, 11292-11298.

50. A. C. Thenuwara, S. L. Shumlas, N. H. Attanayake, Y. V. Aulin, I. G. McKendry, Q. Qiao, Y. M. Zhu, E. Borguet, M. J. Zdilla and D. R. Strongin, ACS Catal., 2016, 6, 7739-7743.

51. J. W. Su, G. L. Xia, R. Li, Y. Yang, J. T. Chen, R. H. Shi, P. Jiang and Q. W. Chen, J. Mater. Chem. A, 2016, 4, 9204-9212.

52. M. D. Merrill and R. C. Dougherty, J. Phy. Chem. C, 2008, 112, 3655-3666.53. Y. W. Liu, H. Cheng, M. J. Lyu, S. J. Fan, Q. H. Liu, W. S. Zhang, Y. D. Zhi, C. M.

Wang, C. Xiao, S. Q. Wei, B. J. Ye and Y. Xie, J. Am. Chem. Soc., 2014, 136, 15670-15675.

54. F. Song and X. L. Hu, J. Am. Chem. Soc., 2014, 136, 16481-16484.55. X. Zhang, H. M. Xu, X. X. Li, Y. Y. Li, T. B. Yang and Y. Y. Liang, ACS Catal.,

2016, 6, 580-588.56. H. Hu, B. Y. Guan, B. Y. Xia and X. W. Lou, J. Am. Chem. Soc. , 2015, 137, 5590-

5595.57. P. F. Liu, S. Yang, L. R. Zheng, B. Zhang and H. G. Yang, J. Mater. Chem. A, 2016,

4, 9578-9584.58. Y. Liang, Y. Li, H. Wang, J. Zhou, J. Wang, T. Regier and H. Dai, Nat. Mater., 2011,

10, 780-786.59. Y. Yang, H. L. Fei, G. D. Ruan and J. M. Tour, Adv. Mater. , 2015, 27, 3175-3180.60. Y. G. Li, P. Hasin and Y. Y. Wu, Adv. Mater., 2010, 22, 1926-1929.61. N. Jiang, B. You, M. L. Sheng and Y. J. Sun, ChemCatChem, 2016, 8, 106-112.62. F. Song and X. L. Hu, Nat. Commun., 2014, 5, 4477.63. H. F. Liang, F. Meng, M. Caban-Acevedo, L. S. Li, A. Forticaux, L. C. Xiu, Z. C.

Wang and S. Jin, Nano Lett., 2015, 15, 1421-1427.64. H. Li, C. Tsai, A. L. Koh, L. L. Cai, A. W. Contryman, A. H. Fragapane, J. H. Zhao,

H. S. Han, H. C. Manoharan, F. Abild-Pedersen, J. K. Norskov and X. L. Zheng, Nat. Mater., 2016, 15, 48-53.

65. J. Kibsgaard and T. F. Jaramillo, Angew. Chem. Int. Ed., 2014, 53, 14433-14437.66. Q. Lu, G. S. Hutchings, W. T. Yu, Y. Zhou, R. V. Forest, R. Z. Tao, J. Rosen, B. T.

Yonemoto, Z. Y. Cao, H. M. Zheng, J. Q. Xiao, F. Jiao and J. G. G. Chen, Nat. Commun., 2015, 6, 6567.

67. Z. P. Shi, Y. X. Wang, H. L. Lin, H. B. Zhang, M. K. Shen, S. H. Xie, Y. H. Zhang, Q. S. Gao and Y. Tang, J. Mater. Chem. A, 2016, 4, 6006-6013.

68. J. R. McKone, B. F. Sadtler, C. A. Werlang, N. S. Lewis and H. B. Gray, ACS Catal., 2013, 3, 166-169.

69. C. C. L. McCrory, S. Jung, I. M. Ferrer, S. M. Chatman, J. C. Peters and T. F. Jaramillo, J. Am. Chem. Soc., 2015, 137, 4347-4357.

70. Y. J. Sun, C. Liu, D. C. Grauer, J. K. Yano, J. R. Long, P. D. Yang and C. J. Chang, J. Am. Chem. Soc., 2013, 135, 17699-17702.

71. C. L. Fan, D. L. Piron, A. Sleb and P. Paradis, J. Electrochem. Soc., 1994, 141, 382-387.

65

72. M. S. Faber, R. Dziedzic, M. A. Lukowski, N. S. Kaiser, Q. Ding and S. Jin, J. Am. Chem. Soc., 2014, 136, 10053-10061.

73. H. W. Liang, S. Bruller, R. H. Dong, J. Zhang, X. L. Feng and K. Mullen, Nat. Commun., 2015, 6, 7992.

74. L. Liao, S. N. Wang, J. J. Xiao, X. J. Bian, Y. H. Zhang, M. D. Scanlon, X. L. Hu, Y. Tang, B. H. Liu and H. H. Girault, Energy Environ. Sci., 2014, 7, 387-392.

75. D. Voiry, H. Yamaguchi, J. W. Li, R. Silva, D. C. B. Alves, T. Fujita, M. W. Chen, T. Asefa, V. B. Shenoy, G. Eda and M. Chhowalla, Nat. Mater. , 2013, 12, 850-855.

76. H. Vrubel and X. L. Hu, Angew. Chem. Int. Ed., 2012, 51, 12703-12706.77. Z. Y. Yu, Y. Duan, M. R. Gao, C. C. Lang, Y. R. Zheng and S. H. Yu, Chem. Sci. ,

2017, 8, 968-973.78. A. Han, H. Y. Zhang, R. H. Yuan, H. X. Ji and P. W. Du, ACS Appl. Mater.

Interfaces, 2017, 9, 2240-2248.79. S. C. Huang, Y. Y. Meng, S. M. He, A. Goswami, Q. L. Wu, J. H. Li, S. F. Tong, T.

Asefa and M. M. Wu, Adv. Func. Mater., 2017, 27.80. J. Q. Tian, Q. Liu, A. M. Asiri and X. P. Sun, J. Am. Chem. Soc., 2014, 136, 7587-

7590.81. C. L. Xiao, Y. B. Li, X. Y. Lu and C. Zhao, Adv. Func. Mater., 2016, 26, 3515-3523.82. Y. Yu, P. Li, X. F. Wang, W. Y. Gao, Z. X. Shen, Y. N. Zhu, S. L. Yang, W. G. Song

and K. J. Ding, Nanoscale, 2016, 8, 10731-10738.83. X. G. Wang, W. Li, D. H. Xiong and L. F. Liu, J. Mater. Chem. A, 2016, 4, 5639-

5646.84. C. G. Read, J. F. Callejas, C. F. Holder and R. E. Schaak, ACS Appl. Mater.

Interfaces, 2016, 8, 12798-12803.85. X. D. Yan, K. X. Li, L. Lyu, F. Song, J. He, D. M. Niu, L. Liu, X. L. Hu and X. B.

Chen, ACS Appl. Mater. Interfaces, 2016, 8, 3208-3214.86. G. F. Chen, T. Y. Ma, Z. Q. Liu, N. Li, Y. Z. Su, K. Davey and S. Z. Qiao, Adv.

Func. Mater., 2016, 26, 3314-3323.87. X. L. Zhu, C. Tang, H. F. Wang, B. Q. Li, Q. Zhang, C. Y. Li, C. H. Yang and F.

Wei, J. Mater. Chem. A, 2016, 4, 7245-7250.88. L. G. Feng, H. Vrubel, M. Bensimon and X. L. Hu, Phys. Chem. Chem. Phys., 2014,

16, 5917-5921.89. M. Ledendecker, S. K. Calderon, C. Papp, H. P. Steinruck, M. Antonietti and M.

Shalom, Angew. Chem. Int. Ed., 2015, 54, 12361-12365.90. D. N. Liu, Q. Lu, Y. L. Luo, X. P. Sun and A. M. Asiri, Nanoscale, 2015, 7, 15122-

15126.91. A. Han, H. L. Chen, H. Y. Zhang, Z. J. Sun and P. W. Du, J. Mater. Chem. A, 2016,

4, 10195-10202.92. X. D. Jia, Y. F. Zhao, G. B. Chen, L. Shang, R. Shi, X. F. Kang, G. I. N. Waterhouse,

L. Z. Wu, C. H. Tung and T. R. Zhang, Adv. Energy Mater., 2016, 6.93. Y. H. Liang, Q. Liu, Y. L. Luo, X. P. Sun, Y. Q. He and A. M. Asiri, Electrochim.

Acta, 2016, 190, 360-364.94. C. B. Ouyang, X. Wang, C. Wang, X. X. Zhang, J. H. Wu, Z. L. Ma, S. Dou and S.

Y. Wang, Electrochim. Acta, 2015, 174, 297-301.95. W. X. Zhu, X. Y. Yue, W. T. Zhang, S. X. Yu, Y. H. Zhang, J. Wang and J. L. Wang,

Chem. Commun., 2016, 52, 1486-1489.

66

96. D. Ansovini, C. J. J. Lee, C. S. Chua, L. T. Ong, H. R. Tan, W. R. Webb, R. Raja and Y. F. Lim, J. Mater. Chem. A, 2016, 4, 9744-9749.

97. W. Z. Fang, D. N. Liu, Q. Lu, X. P. Sun and A. M. Asiri, Electrochem. Commun., 2016, 63, 60-64.

98. A. Indra, P. W. Menezes, N. R. Sahraie, A. Bergmann, C. Das, M. Tallarida, D. Schmeisser, P. Strasser and M. Driess, J. Am. Chem. Soc. , 2014, 136, 17530-17536.

99. S. A. Eliziario, J. M. de Andrade, S. J. G. Lima, C. A. Paskocimas, L. E. B. Soledade, P. Hammer, E. Longo, A. G. Souza and L. M. G. Santos, Mater. Chem. Phys., 2011, 129, 619-624.

100. D. P. Hashim, N. T. Narayanan, J. M. Romo-Herrera, D. A. Cullen, M. G. Hahm, P. Lezzi, J. R. Suttle, D. Kelkhoff, E. Munoz-Sandoval, S. Ganguli, A. K. Roy, D. J. Smith, R. Vajtai, B. G. Sumpter, V. Meunier, H. Terrones, M. Terrones and P. M. Ajayan, Sci. Rep., 2012, 2.

101. C. W. Ong, H. Huang, B. Zheng, R. W. M. Kwok, Y. Y. Hui and W. M. Lau, J. Appl. Phys., 2004, 95, 3527-3534.

102. P. W. Menezes, A. Indra, N. R. Sahraie, A. Bergmann, P. Strasser and M. Driess, ChemSusChem, 2015, 8, 164-171.

103. B. You and Y. J. Sun, Adv. Energy Mater., 2016, 6, 1502333.104. L. Yu, H. Q. Zhou, J. Y. Sun, F. Qin, F. Yu, J. M. Bao, Y. Yu, S. Chen and Z. F.

Ren, Energy Environ. Sci., 2017, 10, 1820-1827.